I know it records to a hard drive.I was wondering what compression does it use.How many frames per sec and such..I captured video from a vcr to my computer and it uses alot of space compared to a dvr and the quality isn't as good either when compared to a video that was recorded to the dvr
how does a 501 record video??
- Thread starter lime4x4
- Start date
- Latest activity Latest activity:
- Replies 14
- Views 4K
You are recording an analog signal off the VCR but when an E* receiver records to the hard drive it is recording a digital data stream that is already compressed. When the DVR plays it back then it is like the receiver is getting the signal direct from satellite.
Believe me, the 501 uses a lot of space to capture the video. It uses mpeg2 and takes up about 1gb per hour. That's insanely huge compared to the newer codecs of today where the same quality can be achieved in about half that size.
Since I am the CTO (Chief Technology Officer) for a cable/satellite channel (xy.tv), I am have researched alot of this stuff.
Mpeg2 is the standard for pretty much the whole world for broadcast video. Every source available (Terrestrial TV, Satellite, DVD) uses it, and it works quite well. Mpeg4 is a new emerging standard that even the Mpeg group talks about as "the standard for multimedia for the fixed and mobile web" http://www.chiariglione.org/mpeg/ and not a broadcast format.
The problems with changing over from Mpeg2 to Mpeg4 are many:
1) Complete changout of all equipment, from encoding to display
2) Non-proven at broadcast quality like mpeg2 is
3) Very high compression "horsepower" needed
All professional video editing is done typically in either Mpeg2 or DV or some other high quality format. Perhaps Mpeg4 as a delivery method for web based content might be good, but right now using it for HD or SD television is not practical at all.
When one realizes just how much video content they are getting using Mpeg2 compared to an uncompressed video stream, it is quite amazing. Lets compare 1 minute of video using Uncompressed, MiniDV and Mpeg2 at DVD, and Satellite quality:
Uncompressed:
720 (wide)x 480 (high) x 256 (levels) x 3(colors) x 30(frames/sec) x (60 sec)= 477757440000 bits/minute or 59719680000 bytes/minute or around 55 gigabytes/minute for uncompressed video.
MiniDV:
It runs 25 megabits/sec * 60 (seconds) = 1500 megabits/minute or 187 megabytes/minute or .1831 gigabytes/minute
DVD Mpeg2 (6 megabits/second):
6 * 60 = 360 megabits/minute or 45 megabytes/minute or .0439 gigabytes/minute
Satellite Mpeg2 (2.5 megabits/second):
2.5 * 60 = 150 megabits/minute or 18.75 megabytes/minute or .0183 gigabyts/minute
So Mpeg 2 satellite delivered video is only .0332727% of the origional bitrate of uncompressed video.
My point: Mpeg2 does a pretty good job at compressing video.
As to recording video from a 501 to a computer, the best is to use as high a quality setting as you can stand to use. Every time you compress/uncompress/recompress video, you lose picture quality. That is one of the reasons that some of the regular channel on satellite and cable look bad because the bitrate used by the uplinker is too low to handle the recompression by the provider.
Reedl
Mpeg2 is the standard for pretty much the whole world for broadcast video. Every source available (Terrestrial TV, Satellite, DVD) uses it, and it works quite well. Mpeg4 is a new emerging standard that even the Mpeg group talks about as "the standard for multimedia for the fixed and mobile web" http://www.chiariglione.org/mpeg/ and not a broadcast format.
The problems with changing over from Mpeg2 to Mpeg4 are many:
1) Complete changout of all equipment, from encoding to display
2) Non-proven at broadcast quality like mpeg2 is
3) Very high compression "horsepower" needed
All professional video editing is done typically in either Mpeg2 or DV or some other high quality format. Perhaps Mpeg4 as a delivery method for web based content might be good, but right now using it for HD or SD television is not practical at all.
When one realizes just how much video content they are getting using Mpeg2 compared to an uncompressed video stream, it is quite amazing. Lets compare 1 minute of video using Uncompressed, MiniDV and Mpeg2 at DVD, and Satellite quality:
Uncompressed:
720 (wide)x 480 (high) x 256 (levels) x 3(colors) x 30(frames/sec) x (60 sec)= 477757440000 bits/minute or 59719680000 bytes/minute or around 55 gigabytes/minute for uncompressed video.
MiniDV:
It runs 25 megabits/sec * 60 (seconds) = 1500 megabits/minute or 187 megabytes/minute or .1831 gigabytes/minute
DVD Mpeg2 (6 megabits/second):
6 * 60 = 360 megabits/minute or 45 megabytes/minute or .0439 gigabytes/minute
Satellite Mpeg2 (2.5 megabits/second):
2.5 * 60 = 150 megabits/minute or 18.75 megabytes/minute or .0183 gigabyts/minute
So Mpeg 2 satellite delivered video is only .0332727% of the origional bitrate of uncompressed video.
My point: Mpeg2 does a pretty good job at compressing video.
As to recording video from a 501 to a computer, the best is to use as high a quality setting as you can stand to use. Every time you compress/uncompress/recompress video, you lose picture quality. That is one of the reasons that some of the regular channel on satellite and cable look bad because the bitrate used by the uplinker is too low to handle the recompression by the provider.
Reedl
The Dishrip solution has worked for me, you'll use between 1-1.3 GB/hr for recordings off of it.
Then you get to edit the recordings . . .
I had been doing it with TMPGenc, that way I can resize the picture and edit out commercials at the same time, but it's such a pain that I realize there's not a whole lot that I want to keep that bad. (Biggest pain was the time to wait for it to re-encode). ALthough there are other methods.
Used to do that with SG1, but why bother when the DVD sets look way better than what is sent out on Sci-Fi....
Then you get to edit the recordings . . .
I had been doing it with TMPGenc, that way I can resize the picture and edit out commercials at the same time, but it's such a pain that I realize there's not a whole lot that I want to keep that bad. (Biggest pain was the time to wait for it to re-encode). ALthough there are other methods.
Used to do that with SG1, but why bother when the DVD sets look way better than what is sent out on Sci-Fi....
Yeah - SG-1 on SHO was much better than on SciFi. I hate how SciFi (and others) throw crap all over the screen. I've got a lot of SF on tape, and now DVDs are available. Oh well.drcos said:
OT: Shameless plug re: SG-1. Ever hear of Gatecon? It's the #1 SG-1 fan convention. Lots of good times.
reedl said:
Being a video server wrangler, I am a peer of yours and can confirm that what you say is exactly on the money regarding MPEG-2, MPEG-4, the issues surrounding it and the techniques one should use in capturing it.
Might I add that HDTV begins life at 360 Mbps and ends up being transmitted terrestrially at only 19.3 Mbps, so healthy levels of compression there might also reflect a lot of success of MPEG compression techniques.
But this also got me thinking. I have a Panasonic DVD recorder that has four settings (plus a "flex" setting). If I record at "SP", I can get two hours of recording on a disc, and the bit rate is about 4 to 7 Mbps. It can then obviously do a great job of recording DBS signals, which originate at 2.5 Mbps. If I use a lower setting, not only does the bit rate go down, but the entire resolution changes, downrezzing from 704 to 352, which is a noticeable drop in quality and is likely done to prevent bit starving at lower settings. Would it not make sense for a DVDR manufacturer to offer a fixed bit rate setting of 2.5 at the higher resolution, which should be able to get 3 or 4 hours of video on a disk at the same PQ that essentially is displayed by a STB or PVR? If so, that would certainly be a killer feature. I'd appreciate any input folks might have on this.
From what I have seen lately, HDTV is even less that 19.3 Mbps. Considering that Boston's WGBH broadcasts an HD channel, and 2 SD channels, and that I think that the major networks are only doing 11 or so Mbps, there is even better compression going on.TyroneShoes said:
The problem is that when you take a previously compressed signal (like a satellite delivered mpeg video), and uncompress it (convert it to Video), and then recompress it, you get more artifacts that were origionally in the video. You cannot simply reencode the video again at 2.5 and get any sort of great quality.
There was research on MP3's that I read a couple of years ago. It said that if you compress/uncompress a CD quality .WAV file 10 times at 160 kbps (which is higher than most people do with mp3's), you will get only .06% of the original signal in the resultant audio file. I am sure the numbers are the same video video compression.
Reedl
drcos said:
I also use TMPGenc also for important stuff, and having to take all the time to re author also saves me from burning junk to DVD or CDR. If I really want something questionable I'll dump it onto the VCR.
I have not found any video card/software combo as flexible as ATI in allowing you to decide on the ultimate resolution and how much space you want to use to capture and burn digital tv to a computer or DVD. You can easily go MPEG2 or MPEG 1.
The latest versions of ATI MMC gives you canned digital vcr output selections of DVD, Good, Longer, and Video CD recording quality. You can also set custom MPEG2. So, while TMPGenc DVD authoring might balloon it up a bit again, you can really cut down dramatically on file size by reducing desired resolution. And you can easily cut it back big time and put it on CDR's if you want.
http://www.digitalfaq.com/capture/atimpeg/atimpeg.htm
Although I haven't experimented with it yet, I understand the latest ATI cards and software allows export to MPEG4.
reedl said:
19.3 refers to the entire 8VSB bitstream. Inside that you can fit one 1080i plus a 720p with room to spare, or even four 480p's, so yes, the end result for a single HD stream is even less. Yet it begins life at 360.
Whether you get significantly more artifacts depends upon the profile used. You can recompress at high profiles a number of times without significant degradation. Your excellent example of MP3 is the exact opposite...re-encoding using a severe algorithm only twice will indeed give you crappy results. 10 times is just asking for trouble. (I find that AAC at 128 kbps beats MP3 at 192 every time, BTW). The question that remains is if a 2.5 bitstream is decoded and then re-encoded at 2.5, what do you get?
When you create a DVDR from a DBS box you are encoding twice anyway. The encoded signal is of course decoded to NTSC before it reaches your DVDR, where it is encoded a second time (possibly a third if going to the HD first and then to disc at another rate). What I am proposing is to still do exactly that, but to encode at the same rate (on the pass from STB to DVDR) that was used to encode at the uplink center in the first place. My guess would be that it would be nearly as good as SP with much more capacity, yet better than LP, which is pixel-resolved to an unacceptable amount (to my eyes, anyway). The only variable is how much compression is too much when doing this twice. My fear is that since they change the rez at LP, they likely have a good reason for doing it. I found that I can drop to 6 Mbps in the pro arena before I notice the difference, but we use 12 Mb just to be safe.
